[systemd-devel] udev took too long to start

[Lennart Poettering]

On Fri, 20.05.11 17:40, Chen Jie (chenj at lemote.com) wrote:

> 2011/5/19 Chen Jie <chenj at lemote.com>:
> > 2011/5/18 Kay Sievers <kay.sievers at vrfy.org>:
> >>
> >> Completely different! That's an Intel Core Duo 1.4 GHz laptop:
> >>  systemd-analyze blame | grep udev
> >>  87ms udev-trigger.service
> >>  13ms udev.service
> > I updated systemd(to v26) and udev(to 168), still got ~1s startup time
> > of udev.service plus udev-trigger.service.
> >
> > May be some distribution added udev rules cause this? I'll do a
> > bootchart to see the detail.
> See the attachment.

There are a ton of things udev seems to be calling. A lot of it doesn't
look right. i.e. systemctl being called from udev doesn't look
right. And ps? grep? hdparm?? sync??? alsa-utils looks wrong too? mount?

The exim script looks really borked too. And the mkdir/rm sprinkled
around dbus/rsyslog is suspicous too.

I think the distro you are using is a bit too hack-happy... Your
downstream udev maintainers really should spend some time on cleaning up
those udev rules. Upstream udev doesn't ship that nonsense!

Lennart

-- 
Lennart Poettering - Red Hat, Inc.